Setup guide
Check your prices in the Pricing Lab before you go live
Test every possible answer combination at once, catch the pricing mistakes your customers would have found, and fix them without hunting through your questions.
The Pricing Lab runs every possible combination of answers through your estimator and shows you the budget, baseline and premium price for each service in one table. An AI check flags anything that looks wrong, and you can describe a change in plain English and have it staged for your review. Run it before you send anyone to your estimator.
The question we get asked more than any other is "how do I know my pricing is right?". Until this existed, the only answer was to take your own estimator over and over, trying combinations. That is slow, and you will never cover every scenario. There are more combinations than you think.
This is the step most accounts skip. When we audited one franchise brand's nineteen locations, sixteen of them had never changed a single price from the template they launched on. Every estimate those customers saw was somebody else's numbers.
Running it
Open your estimator for editing
From your dashboard, find the estimator and click Edit.
Click the Pricing Lab tab
Along the top of the editor: Questions, Pricing Lab, Content Settings, Financing.
Run the AI analysis
This scans every scenario for prices that look too low, ranges that are unusually wide, services returning $0, and anything else likely to surprise a customer.
If nothing is wrong you get an All Clear. If something is off, it names the specific issue.
What good looks like: All Clear, or a short list of flags you understand and agree with.Read the Price Range Summary
This table is where the real value is. Every service, with three columns: Budget, Baseline and Premium. These are the lowest, middle and highest prices your estimator can produce for that service across every answer combination.
Go down the list and ask one question of each row: would I do that job for the budget figure? If the answer is no, that is your problem row, and you have just found it in seconds rather than in front of a customer.
Drill into a scenario if something looks wrong
Below the summary, every individual test scenario is listed with the exact answer combination that produced it. When a summary figure looks strange, this is where you find out which answer caused it.
Asking the AI to fix it for you
Instead of finding a problem, working out the fix, and hunting for the right answer to edit, you can describe what you want in your own words.
Tell it what you want
At the bottom of the Pricing Lab is a box called Pricing expectations. Two ways to use it:
Describe what feels off, if you are not sure of the fix. For example: "Gutter cleaning feels expensive for my market, especially the largest house size."
Give a direct instruction, if you already know. For example: "Increase the bathroom renovation service by 20%."
Plain English. You do not need to know which answer option is causing it. Click Run Analysis. After the first run the button becomes Recheck Analysis.
Review the suggested changes
You get back a set of AI Suggested Changes, one card each.
Each card names the exact service, question and answer option it wants to change. On the card What it tells you What is changing The exact service, question and answer option, so there is no guessing where it lands Old and new price Current value struck through next to the suggestion, for example $0 becoming $0 to $2,000 Why A plain English reason, such as a percentage producing unrealistic totals Scenario tabs Baseline, Random 1 to 5, Premium. Click through to see how the change behaves before you commit This is also how the typos surface. A finish option set to 2,000% instead of 20% jumps straight out here.
Apply a fix. Nothing goes live yet.
Click Apply fix on a card and the button becomes Applied.
Applying stages the change. It does not publish it. Applying stages the change for review. It is a draft edit sitting in the background. Nothing reaches a customer until you say so. Apply as many as you like and they queue up together.
Review, then publish
A marker appears showing you have work waiting, for example "1 change ready to review", with a Review changes link. Next to the affected question you get Jump to change.
The Jump to change link takes you straight to the staged edit. Open the Questions tab, check the new option text and pricing read the way you want, at a level a customer would understand.
Read it as a customer would before you publish it. When you are happy, hit Save & Publish.
Getting the most out of it
- Run it before you publish. Make it the last thing you do before sending anyone to your estimator. Two minutes here saves an awkward conversation later.
- Trust the summary over the scenarios. The individual scenarios are for debugging. The Price Range Summary is what tells you whether your pricing is sane.
- Re-run after every change. Adjusting one answer can move prices in combinations you were not thinking about.
- Check the budget column hardest. The cheapest number your estimator can produce is the one a customer will hold you to.
Troubleshooting
| What you see | What it means |
|---|---|
| A service showing $0 | An answer option has no price set, or a percentage is being applied to a zero base. The AI flags these, and the scenario list shows which combination causes it. |
| A range far wider than expected | Usually one extreme answer option. Drill into the scenarios to find which. |
| A price wildly too high | Nearly always a percentage typo, for example 2,000% instead of 20%. |
| You applied a fix but the live estimator is unchanged | Applying stages the change. Open Questions, review it, and hit Save & Publish. |
| Your change did not save | Check you published rather than only saving, then hard refresh. If it is embedded, clear your site cache too. |
| The numbers are right but customers say they are wrong | Worth checking your own costs rather than the estimator. If your rates have moved, the estimator is faithfully reporting old inputs. |
Does applying a fix change what customers see straight away?
No. It stages the change for your review. It only goes live when you publish.
How many combinations does it actually test?
It works through the combinations your question structure allows, which is why it catches things manual testing does not. The Price Range Summary is the compressed view of all of it.
Can I use it on a live estimator?
Yes. Running the analysis changes nothing on its own, and applied fixes stay staged until you publish.
Do I have to accept the AI suggestions?
No. They are suggestions on cards. Ignore any you disagree with. You know your market better than it does.
Should franchise locations use this?
Yes, and it is the single most valuable thing a new location can do. Your estimator launches on brand template pricing, which is a starting point rather than your pricing.
Would rather we did this for you? Send us a few of your recent real quotes and our team will configure your pricing to match, free. How it works →
Something in the Lab not behaving? Email [email protected] with your estimator link.